Retrospective audit of the user's own Claude Code sessions. Use whenever the user wants to review, analyze, or reflect on their Claude Code chat history / logs / past sessions to surface recurring patterns, repeated mistakes, instruction violations, workflow friction, missing skills or automation, or suboptimal tech choices — and to decide what to improve. Triggers include "meta-review", "/meta-review", "проанализируй мои чаты/сессии с claude", "ретроспектива работы с агентом/claude code", "что можно улучшить в моём workflow / как я работаю с claude", "где я / агент работает неоптимально", "разбери мои логи claude", "review my claude code sessions", "audit how I use claude", "what patterns show up in my history", "what should I automate". Fans out review subagents across distinct lenses, presents a color-coded prioritized findings list (🔴 important / 🟡 significant / 🟢 minor), asks the user what to act on, applies the chosen changes, and appends one JSONL line per run to a per-project log (.meta-review.jsonl) that records the last-review date so each run resumes from there (defaulting to the past week on first run). Use this skill rather than ad-hoc grepping of session files.
How to install meta-review?
fivol/claude-make-better/meta-review
$npx -y skills add fivol/claude-make-better --skill meta-reviewInstalls into the current project.
›Prefer a prompt? Paste this to your agent
Run `npx skills use "https://github.com/fivol/claude-make-better" --skill "fivol/claude-make-better/meta-review"` and follow the generated skill instructions now. Read its complete output, redirecting it to a temporary file first if necessary. Resolve relative paths from the supporting-files directory it provides.
Use the skills in "https://github.com/fivol/claude-make-better" that are relevant to the current task. Run `npx skills add "https://github.com/fivol/claude-make-better"` and select the relevant skills, then follow their instructions.
Related
6 picksType
- improve-codebase-architectureScan a codebase for deepening opportunities, present them as a visual HTML report, then grill through whichever one you pick.Jul 2026
- domain-modelingBuild and sharpen a project's domain model. Use when the user wants to pin down domain terminology or a ubiquitous language, record an architectural decision,…Jul 2026
- caveman-reviewUltra-compressed code review comments. Cuts noise from PR feedback while preserving the actionable signal. Each comment is one line: location, problem, fix.Jul 2026
- codebase-designShared vocabulary for designing deep modules. Use when the user wants to design or improve a module's interface, find deepening opportunities, decide where a…Jul 2026
- code-reviewReview the changes since a fixed point (commit, branch, tag, or merge-base) along two axes — Standards (does the code follow this repo's documented coding…Jul 2026
- resolving-merge-conflictsUse when you need to resolve an in-progress git merge/rebase conflict.Jul 2026
